Search

Many techniques have been developed for the cardinality estimation problem in data management systems. In this document, we introduce a framework for cardinality estimation of query patterns over property graph databases, which makes it possible to analyze, compare and combine different cardinality estimation approaches. This framework consists of three phases: obtaining a set of estimates for some subqueries, extending this set and finally combining the set into a single cardinality estimate for the query. We show that (parts of) many of the existing cardinality estimation approaches can be used as techniques in one of the phases from our framework. The three phases are loosely coupled, this makes it possible to combine (parts of) current cardinality estimation approaches. We create a graph version of the Join Order Benchmark to perform experiments with different combinations of techniques. The results show that query patterns without property constraints can be accurately estimated using synopses for small patterns. Accurate estimation of query patterns with property constraints require new estimation techniques to be developed that capture correlations between the property constraints and the topology in graph databases.

Automated Machine Learning (AutoML) has been used successfully in settings where the learning task is assumed to be static. In many real-world scenarios, however, the data distribution will evolve over time, and it is yet to be shown whether AutoML techniques can effectively design online pipelines in dynamic environments. This study aims to automate pipeline design for online learning while continuously adapting to data drift. For this purpose, we design an adaptive Online Automated Machine Learning (OAML) system, searching the complete pipeline configuration space of online learners, including preprocessing algorithms and ensembling techniques. This system combines the inherent adaptation capabilities of online learners with the fast automated pipeline (re)optimization capabilities of AutoML. Focusing on optimization techniques that can adapt to evolving objectives, we evaluate asynchronous genetic programming and asynchronous successive halving to optimize these pipelines continually. We experiment on real and artificial data streams with varying types of concept drift to test the performance and adaptation capabilities of the proposed system. The results confirm the utility of OAML over popular online learning algorithms and underscore the benefits of continuous pipeline redesign in the presence of data drift., Comment: 25 pages, 8 figures. Machine Learning S.I.: Automating Data Science (2022)

Physical dynamic networks most commonly consist of interconnections of physical components that can be described by diffusive couplings. These diffusive couplings imply that the cause-effect relationships in the interconnections are symmetric, and therefore, physical dynamic networks can be represented by undirected graphs. This article shows how prediction error identification methods developed for linear time-invariant systems in polynomial form can be configured to consistently identify the parameters and the interconnection structure of diffusively coupled networks. Furthermore, a multistep least squares convex optimization algorithm is developed to solve the nonconvex optimization problem that results from the identification method.

A dynamic pricing problem is difficult due to the highly dynamic environment and unknown demand distributions. In this article, we propose a deep reinforcement learning (DRL) framework, which is a pipeline that automatically defines the DRL components for solving a dynamic pricing problem. The automated DRL pipeline is necessary because the DRL framework can be designed in numerous ways, and manually finding optimal configurations is tedious. The levels of automation make nonexperts capable of using DRL for dynamic pricing. Our DRL pipeline contains three steps of DRL design, including Markov decision process modeling, algorithm selection, and hyperparameter optimization. It starts with transforming available information to state representation and defining reward function using a reward shaping approach. Then, the hyperparameters are tuned using a novel hyperparameter optimization method that integrates Bayesian optimization and the selection operator of the genetic algorithm. We employ our DRL pipeline on reserve price optimization problems in online advertising as a case study. We show that using the DRL configuration obtained by our DRL pipeline, a pricing policy is obtained whose revenue is significantly higher than the benchmark methods. The evaluation is performed by developing a simulation for the real-time bidding environment that makes exploration possible for the reinforcement learning agent.

Lignin is a potential resource for biobased aromatics with applications in the field of fuel additives, resins, and bioplastics. Via a catalytic depolymerization process using supercritical ethanol and a mixed metal oxide catalyst (CuMgAlOx), lignin can be converted into a lignin oil, containing phenolic monomers that are intermediates to the mentioned applications. Herein, we evaluated the viability of this lignin conversion technology through a stage-gate scale-up methodology. Optimization was done with a day-clustered Box-Behnken design to accommodate the large number of experimental runs in which five input factors (temperature, lignin-to-ethanol ratio, catalyst particle size, catalyst concentration, and reaction time) and three output product streams (monomer yield, yield of THF-soluble fragments, and yield of THF-insoluble fragments and char) were considered. Qualitative relationships between the studied process parameters and the product streams were determined based on mass balances and product analyses. Linear mixed models with random intercept were employed to study quantitative relationships between the input factors and the outcomes through maximum likelihood estimation. The response surface methodology study reveals that the selected input factors, together with higher order interactions, are highly significant for the determination of the three response surfaces. The good agreement between the predicted and experimental yield of the three output streams is a validation of the response surface methodology analysis discussed in this contribution.

We present the first parallel algorithms that decide strong and branching bisimilarity in linear time. More precisely, if a transition system has n states, m transitions and $$\vert Act \vert $$ | A c t | action labels, we introduce an algorithm that decides strong bisimilarity in $$\mathcal {O}(n+\vert Act \vert )$$ O ( n + | A c t | ) time on $$\max (n,m)$$ max ( n , m ) processors and an algorithm that decides branching bisimilarity in $$\mathcal {O}(n+\vert Act \vert )$$ O ( n + | A c t | ) time using up to $$\max (n^2,m,\vert Act \vert n)$$ max ( n 2 , m , | A c t | n ) processors.

Social touch is essential to human development and communication. Mediated social touch is suggested as a solution for circumstances where distance prevents skin-to-skin contact. However, past research aimed at demonstrating efficacy of mediated touch in reducing stress and promoting helping have produced mixed findings. These inconsistent findings could possibly be due to insufficient control of contextual factors combined with unnatural interaction scenarios. For example, touch occurs less frequently among strangers and is often accompanied with nonverbal visual cues. We investigated how visual presentation of touch, and interpersonal relationship to the sender influence perception, affective experiences, and autonomic responses the touch evoke. Fifty couples of mixed gender were recruited. A mediated touch was repeatedly applied by either the male partner or male confederate to female participants. The latter witnessed through a webcam as the sender caressed a rubber hand or touchpad to send the touch. Following our hypotheses, touch sent by one's partner was perceived softer and more comforting than stranger touch. The partner's touch also resulted in weaker skin conductance responses, particularly when sent by touching a touchpad. In sum, how a mediated touch is experienced depends both on who is touching, and on how the touch is visually represented.

Ethical considerations, including transparency, play an important role when using artificial intelligence (AI) in education. Explainable AI has been coined as a solution to provide more insight into the inner workings of AI algorithms. However, carefully designed user studies on how to design explanations for AI in education are still limited. The current study aimed to identify the effect of explanations of an automated essay scoring system on students’ trust and motivation. The explanations were designed using a needs-elicitation study with students in combination with guidelines and frameworks of explainable AI. Two types of explanations were tested: full-text global explanations and an accuracy statement. The results showed that both explanations did not have an effect on student trust or motivation compared to no explanations. Interestingly, the grade provided by the system, and especially the difference between the student’s self-estimated grade and the system grade, showed a large influence. Hence, it is important to consider the effects of the outcome of the system (here: grade) when considering the effect of explanations of AI in education.

PurposeBusiness models are increasingly recognized as a concept to support innovation in organizations. The implementation and operation of a new or altered business model involves the (re-)design of an organization's business processes and their successful execution. This study reviews and synthesizes the existing body of literature to guide organizations in systematically moving from a business model design to the implementation and operation of the business model through their underlying business processes.Design/methodology/approachA systematic literature review of the methods that bridge business models and business processes is performed. The selected 34 studies are classified according to the method's characteristics and the support in the design, implementation and operation of business models.FindingsThe results of the systematic review provide an overview of existing methods that organizations can adopt when moving from business model design into the implementation and operation of their business model using processes.Originality/valueThis work provides a comprehensive overview and detailed insight into the existing methods that align business models and business processes. It increases the understanding on how these two concepts can be synthesized to support more effective digital innovation in organizations. Based on the review results, knowledge gaps are identified and an agenda for future research bridging the fields of business models and business processes is proposed.

Downtime of industrial assets such as wind turbines and medical imaging devices comes at a sharp cost. To avoid such downtime costs, companies seek to initiate maintenance just before failure. Unfortunately, this is challenging for the following two reasons: On the one hand, because asset failures are notoriously difficult to predict, even in the presence of real-time monitoring devices which signal early degradation. On the other hand, because the available resources to serve a network of geographically dispersed assets are typically limited. In this paper, we propose a novel model referred to as the dynamic traveling maintainer problem with alerts that incorporates these two challenges and we provide three solution approaches on how to dispatch the limited resources. Namely, we propose: (i) Greedy heuristic approaches that rank assets on urgency, proximity and economic risk; (ii) A novel traveling maintainer heuristic approach that optimizes short-term costs; and (iii) A deep reinforcement learning (DRL) approach that optimizes long-term costs. Each approach has different requirements concerning the available alert information. Experiments with small asset networks show that all methods can approximate the optimal policy when given access to complete condition information. For larger networks, the proposed methods yield competitive policies, with DRL consistently achieving the lowest costs.

This paper proposes a method to include the anisotropic hysteresis characteristics of soft-magnetic laminated steels in the magnetic equivalent circuit (MEC) modeling. The loop-based MEC formulation is improved to handle the nonlinearity of the anisotropic magnetic hysteresis, including the dynamic classical eddy-current and excess fields. The developed MEC model is coupled with both the single-valued B-H curve (SVC) in magnetostatic and the dynamic vector hysteresis model (VHM) in transient analysis. Results with a single elementary MEC element show that an alternating magnetic field in a single direction with a peak value smaller than 300 A/m causes a discrepancy of more than 10% between the magnetic flux densities calculated by the VHM and SVC at 50 and 200 Hz excitation frequencies. Moreover, the proposed modeling technique is verified experimentally using the laminated transformer core of TEAM problem 32. The induced voltage calculated by the MEC model with the VHM demonstrates a good agreement with the measurements, while the MEC model with the SVC calculates inaccurate voltage waveforms. Lastly, the total iron loss dissipated in the transformer's iron core is investigated to verify the proposed technique under different excitation levels and frequencies up to 500 Hz. It is observed that the proposed MEC model with the vector hysteresis characteristics of laminated steels is able to calculate the iron loss accurately, while the conventional single-valued curve method fails to estimate the iron loss.

To increase students’ sense of ownership, the present study incorporated elements of Challenge-based learning (CBL) into a mechanical engineering course. CBL is a desirable pedagogy adopted by many universities of technology. Although the boundaries and constraints of a regular course hinder the application of CBL, some of its benefits can be retained. In this paper, we investigate a second-year mechanical engineering course which aimed to give students autonomy in choosing their modelling projects, to facilitate students’ ownership of their learning processes. The mixed-methods analysis of the intervention reveals no particular benefits of the intervention on pilot students’ ownership, compared to a control group. The qualitative analysis suggests that implicit and explicit factors constrain ownership development, namely the anticipated difficulty level and official constraints for selecting projects. Our findings suggest that, currently, there could be too simple assumptions about how providing students with autonomy in selecting their projects allows for student ownership.

Wearable technology enables collecting continuous in situ data from multiple people in various modalities, which can enhance team research and support, as the dynamic coupling of signals between interacting individuals (i.e., team coordination dynamics) is believed to reflect underlying processes and states of team functioning and performance. We conducted a systematic review on existing literature to evaluate the prospective use of wearable technology in research and practice. Using the IMOI framework as an organizing tool, our review revealed considerable support linking team coordination dynamics in different modalities to team functioning and performance, but also explicated the field’s nascent status.

Shaping modulation formats in multi-dimensional (MD) space is an effective approach to harvest spectral efficiency gains in both the additive white Gaussian noise (AWGN) channel and the optical fiber channel. In the first part of this paper, existing MD geometrically-shaped modulations for fiber optical communications are reviewed. It is shown that large gains can be obtained by exploiting correlation in the dimensions or/and by increasing the cardinality of the modulation format. Practical limitations and challenges are also discussed together with efficient solutions. In the second part, we extend the recently proposed four-dimensional (4D) modulation format family based on the constraint of orthant-symmetry to high spectrum efficiencies up to 10 bit/4D-sym by maximizing generalized mutual information for AWGN channel. Reach increases of up to 25% for a multi-span optical fiber transmission system are reported. Lastly,with the help of a recently introduced nonlinear interference (NLI) model, an optimization for designing nonlinear-tolerant 4D modulation formats is introduced for a single-span optical fiber system. Simulation results show that the proposed NLI model-based 4D modulation format could increase the effective SNRs by 0.25 dB with respect to the AWGN channel-optimal 4D modulation format., Comment: 14 pages, 10 figures, accepted by JLT

In this paper, we extend the work of Pimentel et al. (2015) and propose an adjusted estimator of Kendall's $\tau$ for bivariate zero-inflated count data. We provide achievable lower and upper bounds of our proposed estimator and show its relationship with current literature. In addition, we also suggest an estimator of the achievable bounds, thereby helping practitioners interpret the results while working with real data. The performance of the proposed estimator for Kendall's $\tau$ is unbiased with smaller mean squared errors compared to the unadjusted estimator of Pimentel et al. (2015). Our results also show that the bound estimator can be used when knowledge of the marginal distributions is lacking.

We introduce a behavioral model that effectively predicts auctioneers' reserve choice patterns in English clock auctions across varying reserve price formats (whether reserve prices are set ex-ante or ex-post auction events), number of bidders and the distribution of suppliers' costs. In a two-parameter model, which we call Subjective Conditional Probability (SCP), auctioneers have subjective judgement of conditional probabilities. We theoretically show that the SCP explains two intuitive, but sub-optimal, reserve price setting patterns. It predicts and provides rationalizations for ex-ante reserve prices that decrease in the number of bidders and ex-post reserve prices that increase in the realized auction price. We conduct two experiments; one with a uniform cost distribution and another with a left-skewed cost distribution. We validate the SCP model internally and externally by comparing it to a wide range of models, including reduced form linear regression, risk aversion, anticipated regret model, and subjective probability judgement via in-sample and out-of-sample predictions. We conclude that the SCP has strong external validity across a variety of procurement environments.

In the analysis of binary disease classification, single biomarkers might not have significant discriminating power and multiple biomarkers from a large set of biomarkers should be selected. Numerous approaches exist, but they merely work well for mean differences in biomarkers between cases and controls. Biological processes are however much more heterogeneous, and differences could also occur in other distributional characteristics (e.g. variances, skewness). Many machine learning techniques are better capable of utilizing these higher order distributional differences, sometimes at cost of explainability. In this study we propose quantile based prediction (QBP), a binary classification method that is based on the selection of multiple continuous biomarkers. QBP generates a single score using the tails of the biomarker distributions for cases and controls. This single score can then be evaluated by ROC analysis to investigate its predictive power. The performance of QBP is compared to supervised learning methods using extensive simulation studies, and two case studies: major depression disorder and trisomy. Simultaneously, the classification performance of the existing techniques in relation to each other is assessed. The key strengths of QBP are the opportunity to select relevant biomarkers and the outstanding classification performance in the case biomarkers predominantly show variance differences between cases and controls. When only shifts in means were present in the biomarkers, QBP obtained an inferior performance. Lastly, QBP proved to be unbiased in case of absence of disease relevant biomarkers and outperformed the other methods on the MDD case study. More research is needed to further optimize QBP, since it has several opportunities to improve its performance. Here we wanted to introduce the principle of QBP and show its potential., Comment: 26 pages, 5 figures

This paper investigates the implementation of a community energy storage system (CESS) in a neighborhood consisting of households with flexible and inflexible loads, as well as photovoltaic power generation. The system incorporates overlay services (OS) such as fairness management, increasing the fairness level while maximizing CESS utilization, and reputation management, reducing the impact of households trying to take advantage of the system. Additionally, a credit scheme for households, referred to as the quota system, is introduced, allowing households to only import from the battery the amount of energy they have previously contributed. This quota system includes a way to trade credit with other households, as part of the OS. The case study shows that the introduction of the OS on top of a CESS of 50 kWh results in a drop in peak transformer load from 235.7% to 166.9%, with the peak load at 264.3% without a CESS present. Additionally, it is shown that introducing these OS in a CESS does not significantly impact the mean household cost.

Recommender systems are efficient at predicting users’ current preferences, but how users’ preferences develop over time is still under-explored. In this work, we study the development of users’ musical preferences. Exploring musical preference consistency between short-term and long-term preferences in data from earlier studies, we find that users with higher musical expertise have more consistent preferences for their top-listened artists and tags than those with lower musical expertise, while high expertise users also show more diverse listening behavior. Users typically chose to explore genres that were close to their current preferences, and this effect was stronger for expert users. Based on these findings, we conducted a user study on genre exploration to investigate (1) whether it is possible to nudge users to explore more distant genres, (2) how users’ exploration behavior within a genre is influenced by default recommendation settings that balance personalization with genre representativeness in different ways, and (3) how nudging for exploration increases the perceived helpfulness of recommendations for users to explore. Our results show that users were more likely to select the more distant genres if these genres were nudged by presented at the top of the list, however, users with high musical expertise were less likely to do so. We also find that the more representative default slider, which by default recommended more genre-representative tracks, made users set the slider at a less personalized level. The more representative slider position alone did not promote more exploration of users’ current preferences but when combined with the nudge for distant genres, the more representative slider positions effectively nudged users to explore. Nudging to explore does not necessarily lead to an increase in perceived helpfulness. On the one hand, nudging for exploration improves the perceived helpfulness for exploration by making users explore away from their current preferences. On the other hand, it reduces helpfulness due to lower perceived personalization as users move away from their personalized recommendations. To improve perceived helpfulness, it seems necessary to provide a balanced trade-off between exploration and personalization.

Robots are increasingly used in a variety of work environments, but surprisingly little attention has been paid to how robots change work. In this comparative case study, we explore how robotization changed the work design of order pickers and order packers in eight logistic warehouses. We found that all warehouses robotized tasks based on technological functionality to increase efficiency, which sometimes created jobs consisting of ‘left-over tasks’. Only two warehouses used a bottom-up approach, where employees were involved in the implementation and quality of work was considered important. Although the other warehouses did not, sometimes their work design still benefitted from robotization. The positive effects we identified are reduced physical and cognitive demands and opportunities for upskilling. Warehouses that lack attention to the quality of work may risk ending up with the negative effects for employees, such as simplification and intensification of work, and reduced autonomy. We propose that understanding the consequences of robots on work design supports HR professionals to help managing this transition by both giving relevant input on a strategic level about the importance of work design and advocating for employees and their involvement.

Automatically synthesizing controllers for continuous-state nonlinear stochastic systems, while giving guarantees on the probability of satisfying (infinite-horizon) temporal logic specifications crucially depends on abstractions with a quantified accuracy. For this similarity quantification, approximate stochastic simulation relations are often used. To handle the nonlinearity of the system effectively, we use finite-state abstractions based on piecewise-affine approximations together with tailored simulation relations that leverage the local affine structure. In the end, we synthesize a robust controller for a nonlinear stochastic Van der Pol oscillator.

Cross-selling for solution offerings in the post-deployment phase is under-researched. This study posits that although providers of solution offerings have more opportunities to expand their business through enhanced cross-selling opportunities, three mechanisms determine to what extent cross-selling opportunities materialize: solution modularity, sales-service cooperation, and value assessment tooling. Using data collected from 220 US firms, we test our conceptual model while correcting for potential endogeneity. The findings confirm that two of these mechanisms help firms leverage their solutions in the post-deployment by contributing to customer satisfaction and financial performance. Sales-service cooperation has a strong positive moderating effect, whereas value assessment tooling shows a substitution effect together with solution offering, in driving cross-selling opportunities. We also find diminishing returns of cross-selling opportunities on customer satisfaction, implying that an optimum level of cross-selling opportunities exists.

Accurate state estimates are required for increasingly complex systems, to enable, for example, feedback control. However, available state estimation schemes are not necessarily real-time feasible for certain large-scale systems. Therefore, we develop in this paper, a real-time feasible state-estimation scheme for a class of large-scale systems that approximates the steady state Kalman filter. In particular, we focus on systems where the state-vector is the result of discretizing the spatial domain, as typically seen in Partial Differential Equations. In such cases, the correlation between states in the state-vector often have an intuitive interpretation on the spatial domain, which can be exploited to obtain a significant reduction in computational complexity, while still providing accurate state estimates. We illustrate these strengths of our method through a hyperthermia cancer treatment case study. The results of the case study show significant improvements in the computation time, while simultaneously obtaining good state estimates, when compared to Ensemble Kalman filters and Kalman filters using reduced-order models.

Hypothesis: Core-corona supracolloids can be assembled in aqueous dispersions by controlling the physical interactions between the corona and core colloidal particles. A raspberry corona configuration with full surface coverage of the core can be reached by inducing strong attractive interactions between the individual particles. A controlled partial surface coverage of the core, i.e. strawberry configuration, is however, more difficult to achieve. Supracolloids with different surface coverage ratio exhibit unique and multifunctional surface properties. Experiments: By counterbalancing the multiple physical interactions playing a role during the assembly, the configuration and stability of the assemblies could be fine-tuned over a wide range of concentrations. Supracolloids consisting of polyethylene glycol (PEO)-grafted polymer particles covered by silica nanoparticles were assembled with different configurations, by adjusting the pH and ionic strength of the dispersion, the PEO grafting density and the particles concentration. The self-assembly process and resulting configurations were monitored via cryogenic transmission electron microscopy (Cryo-TEM) and light scattering. Findings: The suitable conditions to assemble supracolloids with partial corona coverage have been established. Stable strawberry supracolloids could be prepared, both for diluted (1 wt%) and concentrated (12 wt%) dispersions. These hybrid supracolloids with well-defined configuration are highly relevant to developing advanced water-borne paints and inks, food dispersions, cosmetic and healthcare products.

On 19 September 2022, the first workshop on AI for Manufacturing (AI4M Workshop) took place at ECML-PKDD, the European Conference on Machine Learning and Principles and Practice for Knowledge Discovery in Databases. The workshop brought together researchers and practitioners, from academia and industry, contributing their perspectives. This special section includes five articles in which Artificial Intelligence methods are used to address real problems in the manufacturing industry, ranging from the supply chain, to production, to quality insurance, and predictive maintenance. In this introduction, we present a high-level overview of the current state of the area: observed trends and the main open challenges. This overview is based on these papers, the keynote presentation, the panel discussion, and the discussion that emerged during the workshop.

Feedback is important to improve writing quality; however, to provide timely and personalized feedback is a time-intensive task. Currently, most literature focuses on providing (human or machine) support on product characteristics, especially after a draft is submitted. However, this does not assist students who struggle during the writing process. Therefore, in this study, we investigate the use of keystroke analysis to predict writing quality throughout the writing process. Keystroke data were analyzed from 126 English as a second language learners performing a timed academic summarization task. Writing quality was measured using participants’ final grade. Based on previous literature, 54 keystroke features were extracted. Correlational analyses were conducted to identify the relationship between keystroke features and writing quality. Next, machine learning models (regression and classification) were used to predict final grade and classify students who might need support at several points during the writing process. The results show that, in contrast to previous work, the relationship between writing quality and keystroke data was rather limited. None of the regression models outperformed the baseline, and the classification models were only slightly better than the majority class baseline (highest AUC = 0.57). In addition, the relationship between keystroke features and writing quality changed throughout the course of the writing process. To conclude, the relationship between keystroke data and writing quality might be less clear than previously posited.

Background: Online and blended learning need an appropriate assessment strategy which ensures academic integrity. During the pandemic, many universities have chosen for online proctoring. Although some earlier examples suggest that online proctoring may reduce cheating, the potential side-effects of proctoring are largely unknown. Objectives: Therefore, this study aims to identify the effects of proctoring on students' self-reported temptation to cheat and potentially undesirable side-effects, including test anxiety, perceived exam difficulty, and performance. In addition, we examine which contextual and student characteristics affect test anxiety during online exams. Methods: For this, we collected four waves of survey data throughout a year of teaching during the pandemic at one faculty of a Dutch university, resulting in a total sample of 1760 students within 105 courses. Results and Conclusions: Multi-level analyses showed that while proctoring had no effect on the temptation to cheat, exam difficulty or performance, students reported higher levels of test anxiety. Some learning strategies, internet literacy, access to a reliable technology and a dedicated study space as well as gender and financial stress affect their test anxiety. To conclude, the decision to use online proctoring needs to consider the undesirable side-effect on test anxiety. We discuss practical implications for university administrators, educational designers and teachers to reduce test anxiety.

The energy efficiency of an Internet-of-Things (IoT) receiver can be improved by introducing an adjustable trade-off between signal-quality and energy consumption. In good channel conditions, the receiver can be set to consume less energy per bit, without compromising signal quality in bad channel conditions. We propose a system-level receiver design that enables adequate configuration and combination of signal-quality and energy trade-offs in multiple receiver components. Co-design of all components is essential. We identify the most energy-efficient configurations in our system-level design under different channel conditions. With those configurations, the proposed receiver outperforms a state-of-the-art adjustable receiver with only an adjustable analog front end by several tens of percent in energy per successfully received bit and by 2x in energy-sensitivity configuration range. To show the efficacy of the proposed approach, we integrate a model of the proposed design into the OMNeT++ simulator and show the benefits on an environmental monitoring scenario. In this scenario, we report up to 6x energy savings for the entire transceiver compared to the conventional transceiver design without adjustable receiver.

The evolution of online social networks is highly dependent on the recommended links. Most of the existing works focus on predicting intra-community links efficiently. However, it is equally important to predict inter-community links with high accuracy for diversifying a network. In this work, we propose a link prediction method, called HM-EIICT, that considers both the similarity of nodes and their community information to predict both kinds of links, intra-community links as well as inter-community links, with higher accuracy. The proposed framework is built on the concept that the connection likelihood between two given nodes differs for inter-community and intra-community node-pairs. The performance of the proposed methods is evaluated using link prediction accuracy and network modularity reduction. The results are studied on real-world networks and show the effectiveness of the proposed method as compared to the baselines. The experiments suggest that the inter-community links can be predicted with a higher accuracy using community information extracted from the network topology, and the proposed framework outperforms several measures especially proposed for community-based link prediction. The paper is concluded with open research directions.

Service robots are taking over the frontline. They can possess three types of artificial intelligence (AI): mechanical, thinking, and feeling AI. Although these intelligences determine how service robots can help customers, not much is known about how customers respond to robots of different intelligence. This paper addresses this gap, builds on the appraisal theory of emotions, and employs three online experiments and one field study to demonstrate that customers have different emotional responses to the three types of AI. Particularly, the influence of AI on positive emotions becomes stronger as the AI type becomes more sophisticated. That is, feeling AI relates more strongly to positive emotions than mechanical AI. Also, feeling AI and thinking AI increase spending and loyalty intention through customers’ positive emotions. We also identify important contingency effects of service tiers: mechanical AI is more suitable for low-cost firms, whereas feeling AI mainly benefits full-service providers. Remarkably, none of the three intelligences are directly related to negative emotions; perceived robot autonomy is an important mediator in these relationships. The findings yield concrete managerial guidance as to how smart a service robot should be by pinpointing the right type of AI given the market segment of the service provider.

Faster than Nyquist signaling increases the spectral efficiency of pulse amplitude modulation by accepting intersymbol interference, where an equalizer is needed at the receiver. Since the complexity of an optimal equalizer increases exponentially with the number of the interfering symbols, practical truncated equalizers assume shorter memory. The power of the resulting residual interference depends on the transmit filter and limits the performance of truncated equalizers. In this paper, we use numerical optimizations and the prolate spheroidal wave functions to find optimal time-limited pulses that achieve minimum residual interference. Compared to root raised cosine pulses, the new pulses decrease the residual interference by an order of magnitude, for example, a decrease by 32 dB is achieved for an equalizer that considers four interfering symbols at 57% faster transmissions. As a proof of concept, for the 57% faster transmissions of binary symbols, we showed that using the new pulse with a 4-state equalizer has better bit error rate performance compared to using a root raised cosine pulse with a 128-state equalizer., Comment: 5 pages, 6 figures

Horizontal price agreements can fall within the scope of exemptions to antitrust competition if they are expected to create pro-consumer benefits. Inspired by such horizontal agreements, we introduce a cooperative game in which a set of transport operators can collectively decide at what price to offer sustainable urban mobility services to a pool of travelers. The travelers choose amongst the mobility services according to a multinomial logit model, and the operators aim at maximizing their joint profit under a constant market share constraint. After showing that various well-known allocation rules (i.e., proportional rules and the Shapley value) do not always generate core allocations, we present a core-guaranteeing allocation rule, the market share exchange rule. This rule first allocates to each transport operator the profit he or she generates under collaboration, and then subsequently compensates those transport operators that lose part of their market share, which is paid by the ones that receive some extra market share. This exchange of market share is facilitated by a unique price, which can be expressed as the additional return by cooperating per unit of market share. Finally, we show that, under some natural conditions, the market share exchange rule still sustains the collaboration when the transport operators need to pay back part of the joint profit to society.

Bergstra and Klop have shown that bisimilarity has a finite equational axiomatisation over ACP/CCS extended with the binary left and communication merge operators. Moller proved that auxiliary operators are necessary to obtain a finite axiomatisation of bisimilarity over CCS, and Aceto et al. showed that this remains true when Hennessy’s merge is added to that language. These results raise the question of whether there is one auxiliary binary operator whose addition to CCS leads to a finite axiomatisation of bisimilarity. We contribute to answering this question in the simplified setting of the recursion-, relabelling-, and restriction-free fragment of CCS. We formulate three natural assumptions pertaining to the operational semantics of auxiliary operators and their relationship to parallel composition and prove that an auxiliary binary operator facilitating a finite axiomatisation of bisimilarity in the simplified setting cannot satisfy all three assumptions.

Federated Learning is a distributed machine learning paradigm dealing with decentralized and personal datasets. Since data reside on devices such as smartphones and virtual assistants, labeling is entrusted to the clients or labels are extracted in an automated way. Specifically, in the case of audio data, acquiring semantic annotations can be prohibitively expensive and time-consuming. As a result, an abundance of audio data remains unlabeled and unexploited on users’ devices. Most existing federated learning approaches focus on supervised learning without harnessing the unlabeled data. In this work, we study the problem of semi-supervised learning of audio models via self-training in conjunction with federated learning. We propose FedSTAR to exploit large-scale on-device unlabeled data to improve the generalization of audio recognition models. We further demonstrate that self-supervised pre-trained models can accelerate the training of on-device models, significantly improving convergence within fewer training rounds. We conduct experiments on diverse public audio classification datasets and investigate the performance of our models under varying percentages of labeled and unlabeled data. Notably, we show that with as little as 3% labeled data available, FedSTAR on average can improve the recognition rate by 13.28% compared to the fully supervised federated model.

We introduce the concept of codeterminantal graphs, which generalize the concepts of cospectral and coinvariant graphs. To do this, we investigate the relationship of the spectrum and the Smith normal form (SNF) with the determinantal ideals. We establish a necessary and sufficient condition for graphs to be codeterminantal on R[x], and we present some computational results on codeterminantal graphs up to 9 vertices. Finally, we show that complete graphs and star graphs are determined by the SNF of its distance Laplacian matrix.

In this article, we present a novel linear model predictive control (MPC) strategy for controlling uncertain dynamical systems. The moment based MPC strategy is based on the (centralized) moments of the stochastic cost and constraint functions. We show that moment based MPC formulations yield computationally tractable predictive control problems, due to the explicit analytic expressions of the moments and linearity in the dynamics. This article considers (i) the output regulation MPC configuration, where the output is steered towards a reference and state vector is not correctly accessible to the controller at the decision instants; (ii) the moment based MPC formulation with uncertainties that are non-Gaussian random variables, and (iii) bound and polytopic constraints in the uncertain MPC problem subject to either Gaussian disturbances or uncertainties with symmetric distributions. We demonstrate the effectiveness of the proposed MPC technique on simulation examples such as a double integrator system and a quadruple tank system.

Control over the assembly and morphology of nanoscale functional building blocks is of great importance to hybrid and porous nanomaterials. In this paper, by combining different types of spherical nanoparticles with different size ratios in a hierarchical assembly process which allows us to control the final structure of multi-component assemblies, we discuss self-assembly of an extensive range of supraparticles, labelled as AB particles, and an extension to novel ternary particles, labelled as ABC particles. For supraparticles, the organization of small nanoparticles is known to be inherently related to the size ratio of building blocks. Therefore, we studied the formation of supraparticles prepared by colloidal self-assembly using small silica nanoparticles (SiO2 NPs) attached on the surface of large polystyrene latex nanoparticles (PSL NPs) with a wide size ratio range for complete and partial coverage, by controlling the electrostatic interactions between the organic and inorganic nanoparticles and their concentrations. In this way hierarchically ordered, stable supraparticles, either fully covered or partially covered, were realized. The partially covered, stable AB supraparticles offer the option to create ABC supraparticles of which the fully covered shell contains two different types of nanoparticles. This has been experimentally confirmed using iron oxide (Fe3O4) nanoparticles together with silica nanoparticles as shell particles on polystyrene core particles. Cryo-electron tomography was used to visualize the AB binary and ABC ternary supraparticles and to determine the three-dimensional structural characteristics of supraparticles formed under different conditions.

Skin-compatible printed stretchable conductors that combine a low gauge factor with a high durability over many strain cycles are still a great challenge. Here, a graphene nanoplatelet-based colloidal ink utilizing a skin-compatible thermoplastic polyurethane (TPU) binder with adjustable rheology is developed. Stretchable conductors that remain conductive even under 100% strain and demonstrate high fatigue resistance to cyclic strains of 20-50% are realized via printing on TPU. The sheet resistances of these conductors after drying at 120 °C are as low as 34 ω □ -1mil -1. Furthermore, photonic annealing at several energy levels is used to decrease the sheet resistance to -1mil -1, with stretchability and fatigue resistance being preserved and tunable. The high conductivity, stretchability, and cyclic stability of printed tracks having excellent feature definition in combination with scalable ink production and adjustable rheology bring the high-volume manufacturing of stretchable wearables into scope.

Virtue accounts of innovation ethics have recognized the virtue of creativity as an admirable trait in innovators. However, such accounts have not paid sufficient attention to the way creativity functions as a collective phenomenon. We propose a collective virtue account to supplement existing virtue accounts. We base our account on Kieran’s definition of creativity as a virtue and distinguish three components in it: creative output, mastery and intrinsic motivation. We argue that all of these components can meaningfully be attributed to innovation groups. This means that we can also attribute the virtue of creativity to group agents involved in innovation. Recognizing creativity as a collective virtue in innovation is important because it allows for a more accurate evaluation of how successful innovation generally happens. The innovator who takes a collective virtue account of creativity seriously will give attention to the facilitation of an environment where the group can flourish collectively, rather than only nurturing the individual genius.

Set systems can be visualized in various ways. An important distinction between techniques is whether the elements have a spatial location that is to be used for the visualization; for example, the elements are cities on a map. Strictly adhering to such location may severely limit the visualization and force overlay, intersections and other forms of clutter. On the other hand, completely ignoring the spatial dimension omits information and may hide spatial patterns in the data. We study layouts for set systems (or hypergraphs) in which spatial locations are displaced onto concentric circles or a grid, to obtain schematic set visualizations. We investigate the tractability of the underlying algorithmic problems adopting different optimization criteria (e.g. crossings or bends) for the layout structure, also known as the support of the hypergraph. Furthermore, we describe a simulated-annealing approach to heuristically optimize a combination of such criteria. Using this method in computational experiments, we explore the trade-offs and dependencies between criteria for computing high-quality schematic set visualizations.

The current studies relating to pyramid schemes are mostly about qualitative analysis, whereas the quantitative analysis is still rare owing to the insufficiency in knowledge of their specific trading modes. Often, the trading modes of pyramid schemes are inconspicuous in financial data, making it difficult to be identified in the data. In this study, we propose a quantitative framework for mining trading patterns of pyramid schemes from financial time series data. The framework includes two parts: Long Range Sequence De-noising (LoRSD) algorithm and Contrast Trading Pattern Mining (Contrast TPM) algorithm. LoRSD distinguishes noise items by folding the statistical frequent items and removes the infrequent items recursively. In Contrast TPM, we first identify the frequent one-itemset by comparing the pyramid-related samples with the general samples. Subsequently, a random model is added in the comparative analysis to generate the frequency conditions for mining pyramid scheme patterns. Instead of setting user-defined support thresholds, we adopt contrastive samples as benchmarks in determining the frequency conditions. Our extensive experiments on the financial data set including behaviour of a real-world pyramid scheme demonstrate the effectiveness of our framework in sequence de-noising and mining trading patterns of pyramid schemes from financial time series data.

Previous work has demonstrated that structural forms of empowerment tend to enhance individual and team resilience. However, there is hardly any knowledge about how structural empowerment affects organizational resilience. Moreover, a widespread (though largely untested) assumption is that, in adverse times, power and authority need to be centralized at the top to enhance organizational resilience. This paper explores the effects of empowerment on organizational resilience in an in-depth case study of a Dutch home care organization, in which employees are structurally empowered. The findings from this case study suggest that structural empowerment positively affects organizational resilience, but that this effect is contingent upon a climate of psychological safety as well as top management’s sustained commitment to structural empowerment. We move beyond the extant conceptualization of psychological safety by demonstrating its inter-level nature in the context of structural empowerment, which operates across organizational levels when employees also engage in discussions on tactical and strategic issues. Overall, this study provides an in-depth understanding of how organizations can enhance their resilience by empowering their members, thus also challenging the common wisdom about centralizing power in adverse times.

Stochastic approximation methods for variational inference have recently gained popularity in the probabilistic programming community since these methods are amenable to automation and allow online, scalable, and universal approximate Bayesian inference. Unfortunately, common Probabilistic Programming Languages (PPLs) with stochastic approximation engines lack the efficiency of message passing-based inference algorithms with deterministic update rules such as Belief Propagation (BP) and Variational Message Passing (VMP). Still, Stochastic Variational Inference (SVI) and Conjugate-Computation Variational Inference (CVI) provide principled methods to integrate fast deterministic inference techniques with broadly applicable stochastic approximate inference. Unfortunately, implementation of SVI and CVI necessitates manually driven variational update rules, which does not yet exist in most PPLs. In this paper, we cast SVI and CVI explicitly in a message passing-based inference context. We provide an implementation for SVI and CVI in ForneyLab, which is an automated message passing-based probabilistic programming package in the open source Julia language. Through a number of experiments, we demonstrate how SVI and CVI extends the automated inference capabilities of message passing-based probabilistic programming.

With the growth in complexity of real-time embedded systems, there is an increasing need for tools and techniques to understand and compare the observed runtime behavior of a system with the expected one. Since many real-time applications require periodic interactions with the environment, one of the fundamental problems in guaranteeing their temporal correctness is to be able to infer the periodicity of certain events in the system. The practicability of a period inference tool, however, depends on both its accuracy and robustness (also its resilience) against noise in the output trace of the system, e.g., when the system trace is impacted by the presence of aperiodic tasks, release jitters, and runtime variations in the execution time of the tasks. This work (i) presents the first period inference framework that uses regression-based machine-learning (RBML) methods, and (ii) thoroughly investigates the accuracy and robustness of different families of RBML methods in the presence of uncertainties in the system parameters. We show, on both synthetically generated traces and traces from actual systems, that our solutions can reduce the error of period estimation by two to three orders of magnitudes w.r.t. the state of the art.

Optimal alignments are the basis of conformance checking. For long, researchers have been devoted to the efficiency issue of computing optimal alignments. This paper focuses on the optimality issue. Specifically, we aim to find alignments with maximum synchronous moves and minimum deviations. This paper introduces a coordinate-plane search space, which allows enumerating all the possible alignments. The alignments with maximum synchronous moves are translated into the lowest-cost paths, such that heuristic strategies (such as the Dijkstra algorithm) can be applied. Both theoretical proof and experimental results show that 100% optimality can be achieved.

Direct detection optical orthogonal frequency division multiplexing (DDO-OFDM) systems have been proposed as a well-suited solution for cost-sensitive applications, but also for metro/regional optical links. In this article, we experimentally demonstrate tolerance enhancements of constant-envelope OFDM (CE-OFDM) signals towards Mach–Zehnder modulator (MZM) intermodulation impairments and fiber nonlinear Kerr effects. The additional tolerance is achieved by the reduced peak-to-average power ratio (PAPR) of CE-OFDM signals, unlike conventional OFDM signals that inherently possess high PAPR values and therefore, are prone to distortions induced by the above-mentioned impairments. The experiments show that the transmission of CE-OFDM signals with PAPR =3 dB over 1000 km of dispersion-uncompensated standard single-mode fiber (SSMF) is successfully accomplished. The proper choice of the MZM bias point, as well as the electrical phase modulation index, provides link length enlargements at the same bit error rate. An enlargement around 320 km of SSMF was achieved with an optical launch power of 10 dBm and with a guard-band of ≈16% of the signal bandwidth when compared to conventional DDO-OFDM systems.

The short rise time observed in the PWM voltages generated by ultrafast wide bandgap devices increases the amplitude of voltage harmonics at higher frequencies. These harmonics can excite the resonances of medium-frequency transformers (MFTs), resulting in overvoltages inside the windings during continuous operation. Without further measures, these overvoltages can lead to unexpectedly high electric fields in the insulation material, which can result in partial discharge, accelerated aging, and possible failure of the MFT. To avoid these effects, the mechanism causing the overvoltages has to be understood and quantified during the design process. Based on this, the MFT can be designed in a way that the overvoltages vanish or are tolerable. Therefore, the voltage distribution inside the MFT windings is analyzed by a fully coupled multiconductor transmission line model, which includes the damping effect of electromagnetic losses in the litz wire and the core. This method is verified by measuring the transfer functions of the voltage to ground of individual turns and their voltage waveforms during continuous operation. The waveforms indicate repeating overvoltages inside the windings. A guideline for the design verification and a simplified approach to speed-up the modeling process are presented.

This paper presents the outcomes of a study of the summation of supraharmonic (SH) currents from electric vehicle fast-charging stations (FCS), based on a field measurement in an installation with four 350 kW FCSs. FCSs can contribute to SH emission, between 2 and 150 kHz, in the network and their impact on the grid is quantified by the summation of SH components. Little is known about the summation of SHs in an installation with a power of hundreds of kW, therefore this article aims to contribute to the understanding of the SH current emission and their summation in practice. The results show that the total SH current from the installation increases with the number of active chargers, but not monotonous and that the emission is not constant over time. A diversity factor is proposed and the relation between total and summed current is analyzed to provide further insight into the role of absorption by other chargers. The findings are a step towards modeling the SH summation for FCSs and other installations with large power-electronic devices.

With the rise of deep learning (DL), our world braces for artificial intelligence (AI) in every edge device, creating an urgent need for edge-AI SoCs. This SoC hardware needs to support high throughput, reliable and secure AI processing at ultra-low power (ULP), with a very short time to market. With its strong legacy in edge solutions and open processing platforms, the EU is well-positioned to become a leader in this SoC market. However, this requires AI edge processing to become at least 100 times more energy-efficient, while offering sufficient flexibility and scalability to deal with AI as a fast-moving target. Since the design space of these complex SoCs is huge, advanced tooling is needed to make their design tractable. The CONVOLVE project (currently in Inital stage) addresses these roadblocks. It takes a holistic approach with innovations at all levels of the design hierarchy. Starting with an overview of SOTA DL processing support and our project methodology, this paper presents 8 important design choices largely impacting the energy efficiency and flexibility of DL hardware. Finding good solutions is key to making smart-edge computing a reality.

Urban traffic attributed to commercial and industrial transportation is observed to largely affect living standards in cities due to external effects pertaining to pollution and congestion. In order to counter this, smart cities deploy technological tools to achieve sustainability. Such tools include Digital Twins (DT)s which are virtual replicas of real-life physical systems that control the systems. Research points that DTs can be very beneficial in how they control a physical system by constantly optimizing its performance. The concept has been extensively studied in other technology-driven industries like manufacturing. However, little work has been done with regards to their application in urban logistics. In this paper, we seek to provide a framework by which DTs could be easily adapted to urban logistics networks. To do this, we provide a characterization of key factors in urban logistics for dynamic-decision making. We also survey previous research on DT applications in urban logistics as we found that a holistic overview is lacking. Using this knowledge in combination with the characterization, we produce a conceptual model that describes the ontology, learning capabilities and optimization prowess of an urban logistics digital twin through its quantitative models. We finish off with a discussion on potential research benefits and limitations based on previous research and our practical experience.